The Software Engineer, TPU Host Networking role at Google focuses on the design, development, testing, and deployment of the TPU networking stack. This position is critical for enabling large-scale training and low-latency inference applications using Tensor Processing Units. Engineers in this team will work on optimizing networking performance and enhancing telemetry for better visibility into network behavior. The role requires versatility and a proactive approach to solving complex problems in a fast-paced environment.

• Bachelor's degree or equivalent practical experience.
• 2 years of experience with software development or 1 year of experience with an advanced degree in an industry setting.
• 2 years of experience with developing large-scale infrastructure, distributed systems or networks, or experience with compute technologies, storage or hardware architecture.
• 2 years of experience with networking protocols and troubleshooting.
• 2 years of experience working in C++.

• Write product or system development code.
• Design, develop, test and deploy the TPU networking stack.
• Perform full-stack cross-layer optimization of TPU networking performance for a variety of ML workloads.
• Analyze and debug TPU networking performance issues in production.
• Develop and enhance telemetry to provide deep visibility into network behavior and accelerate troubleshooting.

• Employees at Google are often offered benefits like comprehensive health insurance, 401(k) matching, and flexible work arrangements, among other benefits.